Dries Landuyt

A review of Bayesian belief networks in ecosystem service modelling

A wide range of quantitative and qualitative modelling research on ecosystem services (ESS) has recently been conducted. The available models range between elementary, indicator-based models and complex process-based systems. A semi-quantitative modelling approach that has recently gained importance in ecological modelling is Bayesian belief networks (BBNs). Due to their high transparency, the possibility to combine empirical data with expert knowledge and their explicit treatment of uncertainties, BBNs can make a considerable contribution to the ESS modelling research. However, the number of applications of BBNs in ESS modelling is still limited. This review discusses a number of BBN-based ESS models developed in the last decade. A SWOT analysis highlights the advantages and disadvantages of BBNs in ESS modelling and pinpoints remaining challenges for future research. The existing BBN models are suited to describe, analyse, predict and value ESS. Nevertheless, some weaknesses have to be considered, including poor flexibility of frequently applied software packages, difficulties in eliciting expert knowledge and the inability to model feedback loops. BBNs are increasingly used to analyse, predict and value ecosystem services (ESS).Most BBN applications in ESS modelling target only a single service.Numerous advantages of BBNs in ESS modelling are demonstrated in current applications.Model drawbacks are absence of feedback loops and obligatory variable discretization.Spatially explicit modelling and modelling of ESS bundles are future opportunities.

An ecosystem service approach to support integrated pond management: A case study using Bayesian belief networks – Highlighting opportunities and risks

Freshwater ponds deliver a broad range of ecosystem services (ESS). Taking into account this broad range of services to attain cost-effective ESS delivery is an important challenge facing integrated pond management. To assess the strengths and weaknesses of an ESS approach to support decisions in integrated pond management, we applied it on a small case study in Flanders, Belgium. A Bayesian belief network model was developed to assess ESS delivery under three alternative pond management scenarios: intensive fish farming (IFF), extensive fish farming (EFF) and nature conservation management (NCM). A probabilistic cost-benefit analysis was performed that includes both costs associated with pond management practices and benefits associated with ESS delivery. Whether or not a particular ESS is included in the analysis affects the identification of the most preferable management scenario by the model. Assessing the delivery of a more complete set of ecosystem services tends to shift the results away from intensive management to more biodiversity-oriented management scenarios. The proposed methodology illustrates the potential of Bayesian belief networks. BBNs facilitate knowledge integration and their modular nature encourages future model expansion to more encompassing sets of services. Yet, we also illustrate the key weaknesses of such exercises, being that the choice whether or not to include a particular ecosystem service may determine the suggested optimal management practice.

Combining Biodiversity Resurveys across Regions to Advance Global Change Research

More and more ecologists have started to resurvey communities sampled in earlier decades to determine long-term shifts in community composition and infer the likely drivers of the ecological changes observed. However, to assess the relative importance of and interactions among multiple drivers, joint analyses of resurvey data from many regions spanning large environmental gradients are needed. In this article, we illustrate how combining resurvey data from multiple regions can increase the likelihood of driver orthogonality within the design and show that repeatedly surveying across multiple regions provides higher representativeness and comprehensiveness, allowing us to answer more completely a broader range of questions. We provide general guidelines to aid the implementation of multiregion resurvey databases. In so doing, we aim to encourage resurvey database development across other community types and biomes to advance global environmental change research.

A GIS plug-in for Bayesian belief networks

The complexity and spatial heterogeneity of ecosystem processes driving ecosystem service delivery require spatially explicit models that take into account the different parameters affecting those processes. Current attempts to model ecosystem service delivery on a broad, regional scale often depend on indicator-based approaches that are generally not able to fully capture the complexity of ecosystem processes. Moreover, they do not allow quantification of uncertainty on their predictions. In this paper, we discuss a QGIS plug-in which promotes the use of Bayesian belief networks for regional modelling and mapping of ecosystem service delivery and associated uncertainties. Different types of specific Bayesian belief network output maps, delivered by the plug-in, are discussed and their decision support capacities are evaluated. This plug-in, used in combination with firmly developed Bayesian belief networks, has the potential to add value to current spatial ecosystem service accounting methods. The plug-in can also be used in other research domains dealing with spatial data and uncertainty. Spatial heterogeneity of ES delivery requires spatially explicit accounting methods.Limited availability of primary data promotes the use of knowledge-based BBN models.The proposed GIS BBN plug-in offers a standardized approach to model ES delivery.Diverse probabilistic output maps can be produced to support decision making.The preferred type of output map depends mainly on end-user requirements.

The importance of uncertainties in scenario analyses – A study on future ecosystem service delivery in Flanders

Land use is rapidly changing and is significantly affecting ecosystem service delivery all around the world. The socio-economic context and political choices largely determine land use change. This land use change, driven by socio-economic pressures, will impact diverse elements of the environment including, for example, air quality, soil properties, water infiltration and food and wood production, impacts that can be linked to the provisioning of ecosystem services. To gain more insight into the effects of alternative socio-economic developments on ecosystem service delivery, land use change models are being coupled to ecosystem service delivery models to perform scenario analyses. Although the uncertainty of the results of these kind of scenario analyses are generally far from negligible, studies rarely take them into account. In this study, a cellular automaton land use change model is coupled to Bayesian belief network ecosystem service delivery models to facilitate the study of error propagation in scenario analysis. The proposed approach is applied to model the impact of alternative socio-economic developments on ecosystem service delivery in Flanders, Belgium and to assess the impact of land use allocation uncertainty on the uncertainty associated to future ecosystem service delivery. Results suggest that taking into account uncertainties may have an effect on policy recommendations that come out of the scenario analysis. However, in this study, uncertainties in the applied ecosystem service models were dominant, reducing the importance of accounting for land use allocation uncertainty.

Species distribution models for sustainable ecosystem management

Abstract Reactions to ongoing loss of biodiversity include a variety of restoration actions and are characterised by high costs and uncertainty. Related decision-making can be supported by developing species distribution models (SDMs) that link predictors (both abiotic and biotic) with biotic response variables (e.g., abundance, occurrence, etc.). SDMs can fill in the gaps of current ecological knowledge and predict the potential impact of environmental (including climate) change on species distributions. As climate change already resulted in species shifting their range and an increased risk of extinction, invasion, and disease propagation, SDMs can act as a valuable tool to estimate future species distributions and their effects on ecosystem functioning and related services. Among the variety of modelling techniques used to predict future species distributions, five modelling techniques are selected: decision trees, generalised linear models, artificial neural networks, fuzzy logic, and Bayesian belief networks. The unique advantages of each modelling technique allow the modeller to choose the most appropriate technique in each particular situation. In turn, each modelling technique is characterised by specific drawbacks and is restricted by the limited ecological knowledge related to biotic interactions. Gathering additional ecological knowledge provides the possibility to go beyond simple pattern recognition and to establish more ecologically sound models.

Spatially combining wood production and recreation with biodiversity conservation

Pine plantations established on former heathland are common throughout Western Europe and North America. Such areas can continue to support high biodiversity values of the former heathlands in the more open areas, while simultaneously delivering ecosystem services such as wood production and recreation in the forested areas. Spatially optimizing wood harvest and recreation without threatening the biodiversity values, however, is challenging. Demand for woody biomass is increasing but other pressures on biodiversity including climate change, habitat fragmentation and air pollution are intensifying too. Strategies to spatially optimize different ecosystem services with biodiversity conservation are still underexplored in the research literature. Here we explore optimization scenarios for advancing ecosystem stewardship in a pine plantation in Belgium. Point observations of seven key indicator species were used to estimate habitat suitability using generalized linear models. Based on the habitat suitability and species’ characteristics, the spatially-explicit conservation value of different forested and open patches was determined with the help of a spatially-explicit conservation planning tool. Recreational pressure was quantified by interviewing forest managers and with automated trail counters. The impact of wood production and recreation on the conservation of the indicator species was evaluated. We found trade-offs between biodiversity conservation and both wood production and recreation, but were able to present a final scenario that combines biodiversity conservation with a restricted impact on both services. This case study illustrates that innovative forest management planning can achieve better integration of the delivery of different forest ecosystem services such as wood production and recreation with biodiversity conservation.

Modelling understorey dynamics in temperate forests under global change : challenges and perspectives

The understorey harbours a substantial part of vascular plant diversity in temperate forests and plays an important functional role, affecting ecosystem processes such as nutrient cycling and overstorey regeneration. Global change, however, is putting these understorey communities on trajectories of change, potentially altering and reducing their functioning in the future. Developing mitigation strategies to safeguard the diversity and functioning of temperate forests in the future is challenging and requires improved predictive capacity. Process-based models that predict understorey community composition over time, based on first principles of ecology, have the potential to guide mitigation endeavours but such approaches are rare. Here, we review fourteen understorey modelling approaches that have been proposed during the last three decades. We evaluate their inclusion of mechanisms that are required to predict the impact of global change on understorey communities. We conclude that none of the currently existing models fully accounts for all processes that we deem important based on empirical and experimental evidence. Based on this review, we contend new models are needed to project the complex impacts of global change on forest understoreys. Plant functional traits should be central to such future model developments, as they drive community assembly processes and provide valuable information on the functioning of the understorey. Given the important role of the overstorey, a coupling of understorey models to overstorey models will be essential to predict the impact of global change on understorey composition and structure, and how it will affect the functioning of temperate forests in the future.

Chapter 20 – EBI—An Index for Delivery of Ecosystem Service Bundles∗

The Ecosystem service Bundle Index (EBI) was developed in response to the urgent need for tools that allow rapid and transparent, yet scientific underpinned assessment of ecosystem services. The index is based on a Bayesian network environment in which data on the biophysical conditions and land use properties that drive service delivery are combined to determine the level of service provision. The index points out service optimization opportunities as discrepancies between actual land use and the ecosystem’s biophysical potential. The model can be used for scenario building and offers opportunities to spatially distribute services in a most beneficial way. The EBI was developed as a prototype and tested in a pilot study area using three interacting ecosystem services: carbon sequestration, agricultural production and wood production.

Low probability of a dilution effect for Lyme borreliosis in Belgian forests.

An increasing number of studies have investigated the consequences of biodiversity loss for the occurrence of vector-borne diseases such as Lyme borreliosis, the most common tick-borne disease in the northern hemisphere. As host species differ in their ability to transmit the Lyme borreliosis bacteria Borrelia burgdorferi s.l. to ticks, increased host diversity can decrease disease prevalence by increasing the proportion of dilution hosts, host species that transmit pathogens less efficiently. Previous research shows that Lyme borreliosis risk differs between forest types and suggests that a higher diversity of host species might dilute the contribution of small rodents to infect ticks with B. afzelii, a common Borrelia genospecies. However, empirical evidence for a dilution effect in Europe is largely lacking. We tested the dilution effect hypothesis in 19 Belgian forest stands of different forest types along a diversity gradient. We used empirical data and a Bayesian belief network to investigate the impact of the proportion of dilution hosts on the density of ticks infected with B. afzelii, and identified the key drivers determining the density of infected ticks, which is a measure of human infection risk. Densities of ticks and B. afzelii infection prevalence differed between forest types, but the model indicated that the density of infected ticks is hardly affected by dilution. The most important variables explaining variability in disease risk were related to the density of ticks. Combining empirical data with a model-based approach supported decision making to reduce tick-borne disease risk. We found a low probability of a dilution effect for Lyme borreliosis in a north-western European context. We emphasize that under these circumstances, Lyme borreliosis prevention should rather aim at reducing tick-human contact rate instead of attempting to increase the proportion of dilution hosts.